Evaluation of Induction Warming Stator Cores for Coil Removal

by Dr. Howard W. Penrose Ph.D., CMRP

Evaluation of Induction Warming Stator Cores for Coil Removal

Leo F. Dreisilker
President Dreisilker Electric Motors, Inc. Glen Ellyn, Illinois
AbstractPast methods for warming stators for mechanical coil removal included: gas flame; oven warming; oil bath; and, electric heaters. Independent studies including the Canadian Electrical Associations Evaluation of Electric Motor Repair Procedures, in 1995, and the US Department of the Navy (NAVSEA Motor Repair Manual), identified that these methods have no measureable impact on the core and environment. An induction warming method has been introduced with the intention to improve coil removal times and environmental impact. The purpose of this paper is to compare gas flame to inductive warming and review warming times, post-stripped core condition, and impact on the stator core on integral horsepower machines. Keywords-induction stripping; motor repair; stator core; environment; energy efficiency

Howard W Penrose, Ph.D., CMRP

Vice President, Engineering and Reliability Services Dreisilker Electric Motors, Inc. Glen Ellyn, Illinois methods for increasing the size of electric machines that can be stripped using induction warming have advanced. The newer systems resemble core loss testers with proprietary control strategies and temperature feedback.

I.

INTRODUCTION

The more commonly known method for warming stators as part of a mechanical stripping method is the use of a gas or electric warming table. These tables provide a separation of the heat source and stator core to prevent any uncontrolled temperature. The temperature is observed by the operator using either a contact or infrared thermometer and the readiness for coil removal is determined manually by moving the coil head. The optimal method is to heat the insulating material by warming the core first and allowing the heat to move towards the bore of the stator. The use of an oven is also possible. However, most repair shop ovens are convection types which result in further drying and hardening of the insulation system which makes it more difficult to remove the coils. Coils are removed using a mechanical puller. The preferred type is hydraulic, which results in more control versus pneumatic pullers. There are a variety of pulling methods, such as the one in Figure 1, which all require the coils to be warmed first. Several independent sources site this mechanical approach as being both environmentally sound and having less impact on the core of the electric machine [1][2][3]. Labor times are similar to traditional burnoff oven methods and the total time from start to finish reduced by an average of 7-10 hours [4]. Induction bearing warmers to warm the core and warm the insulation system have been in use since the 1980s for small stators. Within the past several years control logic and

Figure 1.

Hydraulic Stripping Machine

The purpose of this paper is to compare a gas warming stripping method and induction warming method. The comparison is to include times, core loss, type of stator and frame, and general observations. II. THE STUDY Six stators were selected varying in size from three horsepower to 45kW (~60hp), frame sizes from 213T to 324, and from 2 to 6 pole machines. These were first core loss tested and checked for hot spots. They were then put through the induction warming process, monitored with an infrared camera as well as the particular induction warmers thermal control, and the core losses rechecked after cool down. The same stators were then warmed with the gas method, core losses checked and time noted. The core loss tester was a Lexseco Model 2125E, 480 Volt, 270 Amp with 30/60/90 Volt and 2000/2000/1389 Amp outputs with a published accuracy of +/-0.25% of the reading in Watts. The infrared camera was a Fluke Ti30 with a 250oC range. The selected electric motors were as found in Table 1.

978-1-4673-0487-0/12/$31.00 2012 IEEE

214

TABLE I. Stator 1 2 3 4 5 6 HP 50 35 45kW 25 20 3 RPM 3555 1770 4-Pole 1770 1760 1160

MOTOR STATORS FOR WARMING # of Slots 48 48 48 48 48 36 Frame Size 286T 286T Metric 324 256T 213T Ins Class F F F B F B Material Iron Iron Aluminum Iron Steel Iron

The core loss tester is supplied by a main transformer and it was noted that the power supply fluctuated during different times of the day. For the purposes of this paper, the raw data taken from the core loss tester will be presented with error ranges. Ambient temperatures remained between 20oC and 22oC throughout testing. III. INITIAL CORE LOSS TESTS
Figure 3. Infrared Image of Stator During Induction Warming

The commercial core loss tester is set up to produce Watts/Pound (W/lb) measurements. A conversion factor of 1 W/lb to 2.20462262 Watts/kilogram (W/kg) was used to produce W/kg for this paper. Target and actual flux are determined by the instruments software as is the W/lb output.
TABLE II. Stator 1 2 3 4 5 6 Target Flux 652 750 1086 563 345 118 CORE LOSS OF SAMPLES BEFORE WARMING Actual Flux 643 761 1094 564 351 117 W/lb 4.273 3.463 3.548 2.730 2.667 3.482 W/kg 9.420 7.635 7.822 6.019 5.880 7.676 Error in W/kg +/- 0.025 +/- 0.020 +/- 0.020 +/- 0.015 +/- 0.015 +/- 0.020

Table III shows the target temperatures, actual temperature, frame temperature and time to temperature. Table IV shows the post-induction warming core loss using the same set-up as Table II measurements.
TABLE III. Stator 1 2 3 4 5 6 TABLE IV. Stator 1 2 3 4 5 6 INDUCTION WARMING TEMPERATURES AND TIME Actual oC 188 193 190 196 169 188 Frame oC ~188 226 200 ~196 171 197 Time (Minutes) 32 20 90 20 45 15

Target oC 195 200 200 200 200 200

CORE LOSS OF SAMPLES AFTER INDUCTION WARMING Actual Flux 646 765 1091 574 346 119 W/lb 4.203 3.643 3.503 2.770 2.563 3.617 W/kg 9.266 8.031 7.723 6.107 5.650 7.974 Error in W/kg +/- 0.025 +/- 0.020 +/- 0.020 +/- 0.015 +/- 0.015 +/- 0.020

Target Flux 652 750 1086 563 345 118

Figure 2. Test Stator Mounted in Induction Warmer

IV.

INDUCTION WARMING AND CORE LOSSES

It was noted that core losses increased in three of the stators versus the other three. In these cases the controls were set above normal levels in order to see what effect improving the warming time further would have on the stator core losses. It is believed that the rapid temperature increase may have affected the inter-laminar insulation whereas the three normal heating times and settings resulted in improved core losses. Most likely this was caused by de-stressing the cores. A general observation was that additional caution must be taken if hot spots are detected in the laminations during the preheating core loss test. If a significant (>10oC) hot spot is detected, the induction warming method must be applied at lower power levels and monitored closely, if the hot spot cannot be cleared prior to warming. Figure 4 is a post-induction warming infrared of the hot spot test on one of the stators following the core loss test. As noted, no hot spots were identified. A before and after check

The stators were warmed using the induction warming machine. In addition to the built-in sensors, the cores were monitored periodically with an infrared camera in order to observe heat distribution through the core, winding and frame, as shown in Figure 3.

215

on the feet to determine possible stator distortion was also performed and no perceived changes found, indicating that the mechanical properties of the stators were maintained.

value less than, or close to, the original core loss. No hot spots were recorded resulting from this process. VI. REVIEW AND CONCLUSION

Figure 4. Core Loss Test on Post-Induction Warming No Hot Spots

One of the very first production related conclusions that can be noted from the comparison is that the time required to warm a stator for stripping is greatly reduced. However, forcing the warming process faster than recommended methods, such as the under 20 minute warming in this study, can have a negative effect on core loss and, as a result, motor efficiency. Even in these cases the greatest increase noted was 3.9% (Stator #6), well under the no more than 20% allowable increase noted in IEEE Std 1068-2009. Improvements in core losses in the three stators that followed standard warming power levels for the machine used indicate that there is the potential for improvements in efficiency as a result of the methodology. A future study will review the impact of multiple warming processes on a stator using this methodology in order to compare to a Canadian Electrical Association (CEA) [1] study performed in 1994. It was identified during the study that core location measurements are critical in case the stator housing warms enough to allow the stator core to slip. This should be a standard in any stripping process and is worth noting when mechanical stripping methods are used. No stator core movement was identified during the study. While studies, including the previously mentioned CEA study, identify the gas warming method as having no negative environmental impact, the induction warming method eliminates any of the low level emissions that do exist in the older method. Both methods eliminate the larger amounts of green house gasses and particulate emissions from traditional methods such as burn off ovens [5]. Outside of the study stators on machines through 1.1 Megawatt have been warmed for stripping using this method. It is expected that further development will allow continued expansion of the technology to much larger machines. Continued research including a closer look at how quickly a stator may be warmed without negative impact continues, including a correlation between the initial core loss and length of time to warm the stator. REFERENCES
[1] Demand Side Energy Consultants, Inc., Evaluation of Electric Motor Repair Procedures Guidebook, CEA 9205 U 984, Canadian Electrical Association, November, 1995 Naval Sea Systems Command, Technical Manual: Electrical Machinery Repair; Volume 1, Electric Motor Shop Procedures Manual, S6260-BJGTP-010, July 1, 2009 E-Source, Drivepower: Motor Repairs, Rocky Mountain Institute, Colorado, 1993 Penrose, H.W., Anatomy of an Energy Efficient Electric Motor Rewind, IEEE Electrical Insulation Magazine, Vol. 13, Issue 1, pp. 1419, 1997 US EPA, Standards of Performance for New Stationary Sources and Emission Guidelines for Exisitng Sources: Commercial and Industrial Solid Waste Incinerator Units Final Rule, 40 CFR Part 60, Feb. 2011

V.

GAS WARMING METHOD

The gas warming method involves a source of gas, burners, and baffles to direct heated air towards the core of the stator. The stator is mounted vertically to act as a chimney and it warms from the outer part of the core towards the bore and from the side closest to the warming source towards the top. The arrangement is similar regardless of the warming source, other than convection oven heating. For this portion of the study, the stators used for induction warming were re-warmed using this method. The windings remained in the stators, the temperatures were monitored through the warming process, and the times observed. The stators were then core loss tested in order to observe any changes.
TABLE V. Stator 1 2 3 4 5 6 TABLE VI. Stator 1 2 3 4 5 6 Target Flux 652 750 1086 563 345 118 GAS WARMING TEMPERATURES AND TIMES
o

Target C 195 200 200 200 200 200

Actual oC 194 195 199 201 196 195

Frame oC 194 195 199 201 196 195

Time (Minutes) 105 125 85 50 70 35

CORE LOSSES OF STATORS AFTER GAS WARMING Actual Flux 651 759 1068 574 347 117 W/lb 4.263 3.491 3.306 2.848 2.604 3.327 W/kg 9.398 7.696 7.288 6.279 5.741 7.335 Error in W/kg +/- 0.025 +/- 0.020 +/- 0.020 +/- 0.015 +/- 0.015 +/- 0.020

[2]

[3] [4]

[5]

It is noted that the losses have changed again with only Stator #3 continuing to increase in losses. All others had a

216